floating point inconsistency

Cedric Roux cedric.roux@acri-st.fr
Wed Feb 17 14:43:00 GMT 2010


Vincent Lefevre wrote:
> Note that on x86_64 machines, sin and cos are correctly rounded
> (this is not proved, but probably true in practice, assuming no
> bugs in the implementation). I wonder why GCC doesn't use these

well I posted some 32b assembly code a few messages ago that give
different results for "fcos" with the angle of Christoph when run
on ADM or intel CPUs.
So no gcc, no libm, no libc, no lib at all in fact, in the loop.
Just raw CPUs.
And they differ in the last bit for fcos.

So this is not "true in practice."
Maybe in theory, let me check... :-)



More information about the Gcc-help mailing list